Literature reading: Knowledge discovery across documents through concept chain queries

Title: Knowledge discovery across documents through concept chain queries

Authors: Wei Jin and Rohini K. Srihari

Publisher: Sixth IEEE International Conference on Data Mining – Workshops (ICDMW’06), IEEE Computer Society, 2006.

In this paper, the authors combine Srinivasan’s topic profile and closed discovery algorithm with concept chain technique for literature-based hypothesis discovery. The goal is to create, for each semantic type, a ranked list of B terms that link topics A and C together. The ranked list of B terms is represented as a concept chain. Concepts in AP profile are those that co-occur with the topic A within the same sentence. Similarly, concepts in CP profile are those that co-occur with the topic C within the same sentence. All B terms in the BP profile are those that are common between concepts in AP and CP. The proposed technique is applied to the Counterterrorism corpus.

The proposed technique is very similar to the concept chain technique. Only obvious differences are that it is applied to Counterterrorism corpus instead of biomedical domain and Semantex information extraction tool is used rather than MMTx tool. Another drawback is that the technique depends on manual evaluations which are usually affected by biases.

Advertisements
This entry was posted in text mining. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s